R Notebook

library(tidyverse)

Registered S3 method overwritten by 'dplyr':
  method           from
  print.rowwise_df     
Registered S3 methods overwritten by 'dbplyr':
  method         from
  print.tbl_lazy     
  print.tbl_sql      
── Attaching packages ─────────────────────────────────────────────── tidyverse 1.3.0 ──
✓ ggplot2 3.2.1     ✓ purrr   0.3.3
✓ tibble  2.1.3     ✓ dplyr   0.8.4
✓ tidyr   1.0.2     ✓ stringr 1.4.0
✓ readr   1.3.1     ✓ forcats 0.5.0
── Conflicts ────────────────────────────────────────────────── tidyverse_conflicts() ──
x dplyr::filter() masks stats::filter()
x dplyr::lag()    masks stats::lag()

Purrrrrrrrrrrrrrrrrrrrrrr

map() is the main one.

Iterates over a data structure and runs a function that you provide on each “element” of the data structure. replaces ‘for’ loops.

map_() is data type they return e.g. map_dbl() return a vector that contains doubles i.e. numerical

colour_feelings <- list(
    blue   = c("Sad", "Calm"),
    red    = c("Angry", "Energetic", "Warm"),
    green  = c("Calm", "Nature"),
    yellow = c("Happy", "Warm", "Sunny")
)

list on which each element is a character vector.

colour_feelings$blue

[1] "Sad"  "Calm"

for(item in colour_feelings){
  print(length(item))
}

[1] 2
[1] 3
[1] 2
[1] 3

map(.x = colour_feelings, .f = length)

$blue
[1] 2

$red
[1] 3

$green
[1] 2

$yellow
[1] 3

map(.x = colour_feelings, .f = paste, collapse = ", ")

$blue
[1] "Sad, Calm"

$red
[1] "Angry, Energetic, Warm"

$green
[1] "Calm, Nature"

$yellow
[1] "Happy, Warm, Sunny"

colour_translator <- list(
  blue   = "gorm",
    red    = "dearg",   
    green  = "uaine",
    yellow = "buidhe"   
)

We want to paste “Translation:” in front of each of these words

add_translation <- function(text){
  return(paste("Translation: ", text))
}

add_translation("gorm")

[1] "Translation:  gorm"

map(.x = colour_translator, .f = add_translation)

$blue
[1] "Translation:  gorm"

$red
[1] "Translation:  dearg"

$green
[1] "Translation:  uaine"

$yellow
[1] "Translation:  buidhe"

purrr let’s you define little “bespoke” custom-fitted functions to do wrangling.

map(.x = colour_translator, .f = ~ paste("Transalation: ", .x))

$blue
[1] "Transalation:  gorm"

$red
[1] "Transalation:  dearg"

$green
[1] "Transalation:  uaine"

$yellow
[1] "Transalation:  buidhe"

dataframes in R are just lists where each element is a vector and the label of the element is like the column head.

library(CodeClanData)

Attaching package: ‘CodeClanData’

The following object is masked from ‘package:dplyr’:

    starwars

The following object is masked from ‘package:tidyr’:

    population

The following object is masked from ‘package:datasets’:

    volcano

colour_wavelengths <- list(
    blue   = 470,
    red    = 665,
    green  = 550,
    yellow = 600
)

map(colour_translator, nchar)

$blue
[1] 4

$red
[1] 5

$green
[1] 5

$yellow
[1] 6

map(colour_wavelengths, round, digits = -2)

$blue
[1] 500

$red
[1] 700

$green
[1] 600

$yellow
[1] 600

map(colour_wavelengths, ~ .x/(1*(10^9)))

$blue
[1] 4.7e-07

$red
[1] 6.65e-07

$green
[1] 5.5e-07

$yellow
[1] 6e-07

#colour_feelings
map_dbl(colour_feelings, length)

  blue    red  green yellow 
     2      3      2      3 

map_int(colour_feelings, length)

  blue    red  green yellow 
     2      3      2      3 

flatten_chr(colour_feelings)

 [1] "Sad"       "Calm"      "Angry"     "Energetic" "Warm"      "Calm"     
 [7] "Nature"    "Happy"     "Warm"      "Sunny"    

studentslist <- as.list(students)
students

$age_years
 [1] 17 14 14 17 16 17 13 17 16 13 18 17 17 17

$reaction_time
 [1] 0.4420 0.3940 0.5510 0.3550 0.3810 0.6010 0.5940 0.3940 0.4320 0.2990 0.0774
[12] 1.0000 0.4810 0.4500

$text_messages_sent_yesterday
 [1]  45  28   5  35  15 100  31   0 200   0 100 200  35  20

$favorite_school_subject
 [1] "English"                    "Mathematics and statistics"
 [3] "Mathematics and statistics" "Music"                     
 [5] "Mathematics and statistics" "Physical education"        
 [7] "Physical education"         "Other"                     
 [9] "Art"                        "Music"                     
[11] "Computers and technology"   "English"                   
[13] "Mathematics and statistics" "Science"                   

$school_year
 [1] "Year 11" "Year 9"  "Year 8"  "Year 12" "Year 12" "Year 12" "Year 8"  "Year 12"
 [9] "Year 12" "Year 7"  "Year 12" "Year 12" "Year 12" "Year 11"

$height_cm
 [1] 185.0 160.0 152.4 155.0 176.0 177.8 166.0 164.0 162.0 154.0 183.0 157.0 175.3
[14] 173.0

$sleep_hours_schoolnight
 [1]  8.0  7.0  6.0  8.0  5.5  5.0 10.0  7.0  6.0  8.0  8.0  6.0  6.5  6.0

$superpower
 [1] "Telepathy"    "Invisibility" "Telepathy"    "Fly"          "Freeze time" 
 [6] "Invisibility" "Fly"          "Telepathy"    "Telepathy"    "Invisibility"
[11] "Fly"          "Fly"          "Telepathy"    "Telepathy"   

$languages_spoken
 [1] 1.0 2.0 1.0 1.0 2.0 2.0 2.0 1.0 3.0 2.0 2.0 1.0 1.5 2.0

$planned_education_level
 [1] "Graduate degree"       "Graduate degree"       "Graduate degree"      
 [4] "High school"           "Less than high school" "Graduate degree"      
 [7] "Graduate degree"       "Graduate degree"       "Graduate degree"      
[10] "Graduate degree"       "Graduate degree"       "Graduate degree"      
[13] "Graduate degree"       "Graduate degree"      

students <- map_df(students, sort)

conditional mapping:

blue <- list(
    translation = "gorm",
    feelings    = c("Sad", "Calm"),
    primary     = "Yes",
    wavelength  = 470
)

# p is a predicate = logical i.e. condition/question
map_if(.x = blue, .p = is.character, .f = nchar)

$translation
[1] 4

$feelings
[1] 3 4

$primary
[1] 3

$wavelength
[1] 470

# apply paste only to elements of blue with a length > 1

map_if(.x = blue, .p = ~ length(.x) > 1, .f = paste, collapse = ", ")

$translation
[1] "gorm"

$feelings
[1] "Sad, Calm"

$primary
[1] "Yes"

$wavelength
[1] 470

str(colour_list)

List of 4
 $ blue  :List of 4
  ..$ name      : chr "Blue"
  ..$ feelings  : chr [1:2] "Sad" "Calm"
  ..$ primary   : chr "Yes"
  ..$ wavelength: num 470
 $ red   :List of 4
  ..$ name      : chr "Red"
  ..$ feelings  : chr [1:3] "Angry" "Energetic" "Warm"
  ..$ primary   : chr "Yes"
  ..$ wavelength: num 665
 $ green :List of 4
  ..$ name      : chr "Green"
  ..$ feelings  : chr [1:2] "Calm" "Nature"
  ..$ primary   : chr "No"
  ..$ wavelength: num 550
 $ yellow:List of 4
  ..$ name      : chr "Yellow"
  ..$ feelings  : chr [1:3] "Happy" "Warm" "Sunny"
  ..$ primary   : chr "Yes"
  ..$ wavelength: num 600

map(colour_list, "wavelength")

$blue
[1] 470

$red
[1] 665

$green
[1] 550

$yellow
[1] 600

map(colour_list, "feelings")

$blue
[1] "Sad"  "Calm"

$red
[1] "Angry"     "Energetic" "Warm"     

$green
[1] "Calm"   "Nature"

$yellow
[1] "Happy" "Warm"  "Sunny"

not the best way of doing things

map(colour_list, 4)

$blue
[1] 470

$red
[1] 665

$green
[1] 550

$yellow
[1] 600

colour_list_feelings <- map(colour_list, "feelings")
map(colour_list_feelings, length)

$blue
[1] 2

$red
[1] 3

$green
[1] 2

$yellow
[1] 3

API application programming interface APIs return JSON javascript object notation R converts json to nested list

Colours and themes

http://sape.inf.usi.ch/quick-reference/ggplot2/colour

Also look up hexidecimal colours

ggplot(students) +
  aes(x = reaction_time) +
  geom_histogram(fill = "magenta")

ggplot(students) +
  aes(x = reaction_time) +
  geom_histogram(fill = hcl(200, 50, 50))

ggplot(students) +
  aes(x = reaction_time, y = height_cm) +
  geom_point(colour = "violetred1")

ggplot(pets, aes(weight, age, colour = sex)) +
  geom_point()

ggplot(pets, aes(weight, age, colour = sleep)) +
  geom_point() +
  scale_colour_gradient(low = "grey0", high = "grey100")

if colour is fill then scale is fill (make sense?)

ggplot(pets, aes(weight, age, colour = sleep)) +
  geom_point() +
  scale_colour_gradient(low = "seagreen2", high = "turquoise4")

ggplot(pets, aes(weight, age, colour = sleep)) +
  geom_point() +
  scale_colour_gradient2(low = "blue", high = "red", mid = "white", midpoint = 15)

Geom_raster is a heat map

ggplot(volcano, aes(x = x, y = y, fill = height)) +
  geom_raster() +
  scale_fill_gradientn(colours = c("chartreuse1", "maroon1"))

NA

ggplot(volcano, aes(x = x, y = y, fill = height)) +
  geom_raster() +
  scale_fill_gradientn(colours = colorspace::terrain_hcl(10))

https://colorbrewer2.org/#type=sequential&scheme=OrRd&n=3

ggplot(volcano, aes(x = x, y = y, fill = height)) +
  geom_raster() +
  scale_fill_distiller(palette = "OrRd")

ggplot(temp_df) +
  geom_raster(aes(x = month, y = year, fill = max_temp)) +
  scale_fill_gradient(low = "olivedrab2", high = "darkorchid2")

ggplot(temp_df) +
  geom_raster(aes(x = month, y = year, fill = max_temp)) +
  scale_fill_gradient2(low = "dodgerblue4", high = "deeppink4", mid = "darkorange2", midpoint = 10)

ggplot(temp_df) +
  geom_raster(aes(x = month, y = year, fill = max_temp)) +
  scale_fill_gradientn(colours = colorspace::terrain_hcl(28))

ggplot(temp_df) +
  geom_raster(aes(x = month, y = year, fill = max_temp)) +
  scale_fill_distiller(palette = "Dark2")

ggplot(students) +
  aes(x = school_year, fill = superpower) +
  geom_bar()

ggplot(students) +
  aes(x = school_year, fill = superpower) +
  geom_bar() +
scale_fill_hue(h = c(120, 300))

ggplot(students) +
  aes(x = school_year, fill = superpower) +
  geom_bar() +
  scale_fill_brewer(palette = "Set1")

ggplot(students) +
  aes(x = school_year, fill = superpower) +
  geom_bar() +
  scale_fill_grey()

ggplot(students) +
  aes(x = school_year, fill = superpower) +
  geom_bar() +
  scale_fill_grey(start = 0, end = 0.5)

ggplot(students) +
  aes(x = school_year, fill = superpower) +
  geom_bar() +
  scale_fill_manual(values = c(
    "Fly" = "red",
    "Freeze time" = "blue",
    "Invisibility" = "green",
    "Telepathy" = "yellow"
  ))

ggplot(students) +
  aes(x = school_year, fill = superpower) +
  geom_bar() +
  scale_fill_manual(values = c("red", "blue", "green", "yellow"))

ggplot(students) +
  aes(x = school_year, fill = superpower) +
  geom_bar() +
  scale_fill_manual( values = wes_palette("Grandbudapest1"))

Error in wes_palette("Grandbudapest1") : 
  could not find function "wes_palette"

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line()

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line() +
  scale_colour_hue(h = c(60, 2000))

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line() +
  scale_colour_grey(start = 0, end = 0.7)

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line() +
  scale_colour_brewer(palette = "Accent")

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line() +
  scale_colour_manual(values = c("red", "blue", "green", "orange"))

guide_colour_bar for continuous data

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line() +
  scale_colour_hue(guide = guide_legend(nrow = 3))

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line() +
  scale_colour_hue(guide = guide_legend(reverse = TRUE))

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line() +
  scale_colour_hue(guide = guide_legend(keywidth = 1, keyheight = 6, reverse = TRUE))

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line() +
  theme_grey(base_size = 20)

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line() +
  labs(
    title = "Typical Diet of a Chinese Citizen"
  ) +
  theme(
    title = element_text(size = 20, colour = "red", face = "bold")
  )

?theme

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line() +
  labs(
    title = "Typical Diet of a Chinese Citizen"
  ) +
  theme(
    axis.title.x = element_text(size = 20, colour = "red", face = "bold"),
    panel.grid.major = element_line(colour = "black", linetype = "dotted", size = 0.3),
    plot.background = element_rect(fill = "limegreen")
  )

ggplot(chinesemeal) +
  aes(x = Year, colour = FoodType, y = CaloriesPerDay) +
  geom_line() +
  labs(
    title = "Typical Diet of a Chinese Citizen"
  ) +
  theme(
    axis.title.x = element_text(size = 20, colour = "red", face = "bold"),
    panel.grid.major = element_line(colour = "black", linetype = "dotted", size = 0.3),
    plot.background = element_rect(fill = "limegreen"),
    legend.text = element_blank()
  )

install.packages("ggthemes")

trying URL 'https://cran.rstudio.com/bin/macosx/el-capitan/contrib/3.6/ggthemes_4.2.0.tgz'
Content type 'application/x-gzip' length 420298 bytes (410 KB)
==================================================
downloaded 410 KB

The downloaded binary packages are in
    /var/folders/tc/k70prjwj5rs5hjh7zmt604_00000gn/T//RtmpJhNoxO/downloaded_packages

library(ggthemes)

ggplot(scottish_exports) +
  geom_line(aes(x = year, y = exports, colour = sector)) +
  facet_wrap(~sector, scales = 'free_y') +
  theme_excel() +
  theme(
    axis.title.x = element_text(size = 20, colour = "blue", face = "bold"),
    axis.title.y = element_text(size = 20, colour = "blue", face = "bold")
  )

Change order: make into ordered factor

total_sales %>% 
  mutate(branch = factor(branch, levels = c("London", "leeds", "Glasgow", "Leatherhed", "Edinburgh", "Manchester", "Welyn Garden City")))

ggplot(total_sales) +
  aes(x = branch, y = sales) +
  geom_col() +
  coord_flip()

total_sales <- total_sales %>% 
  mutate(branch = fct_reorder(branch, sales))

ggplot(total_sales) +
  aes(x = branch, y = sales) +
  geom_col() +
  coord_flip()